Introduction
The sugarbeet root maggot (SBRM; Tetanops myopaeformis [von Röder]) is the most economically important insect pest of sugarbeet (Beta vulgaris L.) in North America (Hein et al., Reference Hein, Boetel, Godfrey, Harveson, Hanson and Hein2009). Larvae of this pest overwinter in the soil and pupate in the spring. Within a few days, adult flies emerge from the soil, and females migrate to newly planted sugarbeet fields where they deposit eggs at the base of the seedlings. Eggs hatch in 2–3 d, producing larvae that feed on the surface of the sugarbeet root until late July or early August. SBRM feeding can reduce yield by reducing plant stands. However, yield loss is primarily caused by larvae feeding on the root surface throughout the growing season (Hein et al., Reference Hein, Boetel, Godfrey, Harveson, Hanson and Hein2009). In the Red River Valley of Minnesota and North Dakota, yield reductions from SBRM feeding of 10–84% are documented and average 42%, when no SBRM control methods are employed (Campbell et al., Reference Campbell, Anderson, Dregseth and Smith1998).
Sugarbeet germplasm with SBRM resistance became available to the public in 1996 when Campbell et al. (Reference Campbell, Anderson and Dregseth2000) released germplasm line, F1015, with moderate SBRM resistance. SBRM resistance of F1015 was derived from lines of unknown parentage from a discontinued U.S. Department of Agriculture, Agricultural Research Service (USDA-ARS) SBRM-resistance breeding programme in Logan, UT. Greater SBRM resistance became available with the release of F1016 in 1998 (Campbell et al., Reference Campbell, Anderson and Dregseth2000) and F1024 in 2009 (Campbell et al., Reference Campbell, Panella and Smigocki2011). Like F1015, resistance in these lines was derived from the Logan, UT breeding material. In 2016, F1043 (Campbell, Reference Campbell2017) was released which derived resistance from a red globe-shaped Beta accession from Turkey (PI 179180; Callenbach et al., Reference Callenbach, Frye and Anderson1973). F1043 was derived from a cross between PI 179180 and a SBRM-susceptible sugarbeet line from the USDA-ARS's breeding programme in California after selection for SBRM resistance and characteristic sugarbeet root shape and colour.
Genetic resistance provides an economically and environmentally favourable alternative to the current dependence on insecticides. The usefulness of genetic SBRM resistance for elite germplasm breeding programmes is established for F1015, F1016 and F1024. These lines, when used as pollinators, produce hybrids that provide SBRM control comparable to that of commercial insecticides (Campbell and Niehaus, Reference Campbell and Niehaus2008; Campbell et al., Reference Campbell, Miller, Rekoske and Smith2008, Reference Campbell, Panella and Smigocki2011; Campbell, Reference Campbell2018). However, a greater understanding of the relative strengths of SBRM resistance and the genetic relationships between these resistant lines, especially for F1043 which derives resistance from a unique source, would increase the usefulness of resistant germplasm for developing commercial SBRM-resistant hybrids. Therefore, field studies were used to quantify relative resistance of SBRM-resistant germplasm and an analysis of the genetic relationships between SBRM-resistant breeding lines, F1015, F1016, F1024 and F1043, the SBRM-resistant red beet parent of F1043 (PI 179180), and F1055, an unreleased SBRM-resistant population derived from F1024, along with the SBRM-susceptible parents of these lines, were determined using simple sequence repeat (SSR) markers. The purpose of this research was to generate information to assist those using the SBRM-resistant lines as breeding material in choosing the line or lines that best fits their breeding objectives.
Materials and methods
Plant material
Experimental material was comprised of 10 B. vulgaris accessions (Table 1) and included four released SBRM-resistant germplasm lines (F1015, F1016, F1024 and F1043), one population with SBRM resistance (F1055), one SBRM-resistant red beet accession from Turkey (PI 179180) and four SBRM-susceptible sugarbeet germplasm lines or breeding populations that were used as parents in the development of resistant germplasm (C564, F1010, FC709 and FC-CLS). Resistant lines, F1015 (PI 605413) and F1016 (PI 608437), were the products of multiple selection cycles for low SBRM damage within two populations produced by crossing the progeny of F1010 (PI 535818) that was itself subjected to four cycles of mass selection for low SBRM damage, with lines of unknown parentage from a discontinued USDA-ARS SBRM-resistance research programme in Logan, Utah. F1010 was the maternal parent for the population that became F1015. The Utah lines were maternal parents for the population from which F1016 was selected (Campbell, Reference Campbell1990; Campbell et al., Reference Campbell, Anderson and Dregseth2000). F1024 (PI 658654) derived from a cross of F1016 with a SBRM-susceptible line, FC-CLS (Colorado source number: 1996009H2) with resistance to Cercospora leaf spot (CLS; causal agent, Cercospora beticola Sacc.), after multiple selection cycles of the cross's progeny for SBRM resistance (Campbell et al., Reference Campbell, Panella and Smigocki2011). F1055 is an unreleased population (North Dakota source number 13N0009) derived from a cross of F1024 with FC709 (PI 518634), a SBRM-susceptible, Rhizoctonia root and crown rot (causal agent Rhizoctonia solani Kühn) resistant line (Hecker and Ruppel, Reference Hecker and Ruppel1988), after multiple selection cycles of the cross's progeny for SBRM resistance. F1043 was formed by crossing C564 (PI 610317), a SBRM-susceptible line, with PI 179180, a SBRM-resistant accession with red globe-shaped roots originally collected in Turkey, and selection of progeny for SBRM resistance and characteristic sugarbeet root shape and colour (Campbell, Reference Campbell2017). Germplasm lines with C, F or FC in their identifiers are products of USDA-ARS sugarbeet breeding programmes in Salinas, CA, Fargo, ND or Fort Collins, CO, respectively. A summary of the germplasm used in this study is provided in Table 1.
Description of the genetic relationships between SBRM-susceptible and -resistant germplasm used for SSR analysis
USDA germplasm was developed in breeding programmes located in Salinas, CA, Fargo, ND, Fort Collins, CO, and Logan, UT.
United States National Plant Germplasm System (NPGS) plant introduction (PI) numbers are provided where available.
Evaluation of SBRM resistance
SBRM resistance was evaluated under natural infestations in 2013, 2014 and 2015 in field studies conducted near St. Thomas, ND where severe SBRM infestations are an annual occurrence. Experiments were conducted in a randomized complete block design with three replicates, with two 10-m rows, separated by 56 cm, serving as the experimental unit. Plants were grown using standard production practices (Khan, Reference Khan2013) except for no planting-time or postemergence insecticides were applied. In late July or early August, five consecutive roots from the centre of each of the two rows were hand dug, washed and rated on a 0 (no damage) to 9 (more than 75% of the root surface blackened by feeding scars) scale for root maggot damage (Campbell, Reference Campbell, Biancardi, Campbell, Skaracis and Biaggi2005). An average rating for the 10-root sample from each two-row plot was used in the analysis of variance. Significant differences between means were determined using Fisher's protected with Least Significant Difference (LSD), with P ⩽ 0.05.
DNA extraction and SSR analysis
Plants of the 10 germplasm lines were grown in a greenhouse, and leaf tissue was collected individually from three plants from each line or accession. Tissue was flash frozen in liquid nitrogen immediately after collection, lyophilized and ground to a fine powder. Genomic DNA was isolated from lyophilized tissue using a DNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. DNA was quantified spectroscopically (NanoDrop ND-1000, Thermo Scientific, Wilmington, DE, USA). Polymerase chain reaction and genotyping were conducted using SSR forward primers that were appended at the 5′ end with the M13 sequence (5′-CACGACGTTGTAAAACGAC-3′) to allow indirect labelling of reactions. Reverse primers were appended with the sequence GTTTCTT (PIG) at the 5′ end to promote non-templated A addition and to facilitate subsequent genotyping. The M13 universal primer was labelled with a carboxyfluorescein fluorescent tag. SSR allele genotyping was conducted using a carboxy-X-rhodamine (ROX) standard (GeneFlo-625 ROX; CHIMERx, Milwaukee, WI, USA) in an ABI 3730 fluorescent sequencer (POP-6 and a 50-cm array; Applied Biosystems, Foster City, CA, USA). Alleles were scored using GeneMarker software, ver. 1.91 (SoftGenetics, State College, PA, USA).
Statistical analyses
Genetic diversity of the 30 beet genotypes was examined using 22 SSR loci (Fugate et al., Reference Fugate, Fajardo, Schlautman, Ferrareze, Bolton, Campbell, Wiesman and Zalapa2014). Statistics for each of the 22 SSRs were estimated using GenAlEx 6.4 (Peakall and Smouse, Reference Peakall and Smouse2006) and included N (sample size), N a (number of alleles), N e (number of effective alleles), I (information index), H o (observed heterozygosity) and H e (expected heterozygosity). An analysis was also performed comparing the genetic diversity of the SBRM-resistant versus the -susceptible accessions. A database containing allelic information for 22 markers and 30 beet individuals was transformed into a genind object using R statistical software (R Development Core Team, 2011) and the package adegenet 2.1.1 (Jombart, Reference Jombart2008). The function dudi.pca from the package ade4 (Dray and Dufour, Reference Dray and Dufour2007) was used to create a principal coordinates analysis plot (PCoA) with ellipses denoting similarities between data points generated by the package. Using the function aboot from the Poppr package (Kamvar et al., Reference Kamvar, Tabima and Grünwald2014), an unweighted pair group method with arithmetic mean (UPGMA) dendrogram was calculated containing bootstrap support using Prevosti genetic distance (Prevosti et al., Reference Prevosti, Ocana and Alonso1975).
Results and discussion
Field studies were conducted for 3 years to evaluate the relative SBRM resistance of the four released SBRM-resistant lines (F1015, F1016, F1024 and F1043), an unreleased SBRM-resistant population (F1055) and the SBRM-resistant red beet accession (PI 179180) under natural SBRM-infestation conditions (Table 2). Field experiments additionally included F1010, a SBRM-susceptible line which served as parental material in the development of F1015, F1016, F1024 and F1055, and acted as a susceptible control for the study. For all germplasm, in all years, damage ratings were similar for the roots comprising a replicate (data not shown). Damage ratings, therefore, indicated the strength of SBRM resistance for the germplasm line, and was not an average of damage ratings of roots with segregating resistance. Damage ratings for the resistant lines F1016, F1024, F1043 and F1055, and PI 179180 were significantly lower than those of susceptible line F1010 in all years of the study (Table 2). Damage ratings for resistant line F1015 were higher than those for the other SBRM-resistant accessions (F1016, F1024, F1043, F1055 and PI 179180) in all years, but were lower than the susceptible control (F1010) in two of three years. Differences among damage ratings for F1016, F1024, F1043, F1055 and PI 179180 were not significant in any year. In 2014, the only year that included PI 179180 in the study, SBRM damage ratings were nearly equal for F1043 and its parent, PI 179180, suggesting that most, if not all, of the SBRM resistance of PI 179180, an accession with red globe-shaped roots, was captured in the sugarbeet germplasm line, F1043. Overall, these results are consistent with earlier demonstrations of improved SBRM-resistance for germplasm lines F1015, F1016, F1024 and F1043 relative to various SBRM-susceptible lines and adapted hybrids that were first reported when these lines were publicly released (Campbell et al., Reference Campbell, Anderson and Dregseth2000, Reference Campbell, Panella and Smigocki2011; Campbell, Reference Campbell2017).
SBRM damage ratings of five SBRM-resistant germplasm lines, one resistant accession and F1010, a SBRM-susceptible line, from field studies conducted near St. Thomas, ND in 2013–2015
a Damage rating: 0, no feeding scars; 1, one to four small scars; 2, five to ten small scars; 3, up to three large scars or numerous small scars; 4, a few large scars or considerable feeding on lateral roots; 5, several large scars and/or extensive feeding on lateral roots; 6, numerous scars, up to 25% of root blackened with scars; 7, 25–50% of root blackened with scars; 8, 50–75% of root blackened; 9, >75% blackened.
b Not evaluated in 2013 and 2015.
SSR analysis was used to determine genetic relationships between SBRM-resistant germplasm. Parents of SBRM-resistant lines were also included in the analysis to confirm known relationships and to evaluate the effectiveness of the marker analysis. Twenty-two SSR primer-pairs were used to characterize the genetic diversity of three individuals for each accession (Table 3). The 22 microsatellite markers were chosen from 43 polymorphic microsatellite markers that were previously developed from a broad-based sugarbeet transcriptome (Fugate et al., Reference Fugate, Fajardo, Schlautman, Ferrareze, Bolton, Campbell, Wiesman and Zalapa2014). The advantage of using transcript-based SSRs is their higher probability of being associated with functional genes than genomic-based loci (Zalapa et al., Reference Zalapa, Cuevas, Zhu, Steffan, Senalik, Zeldin, McCown, Harbut and Simon2012). The 43 markers were previously validated by their ability to differentiate and detect genetic diversity between eight divergent B. vulgaris accessions (Fugate et al., Reference Fugate, Fajardo, Schlautman, Ferrareze, Bolton, Campbell, Wiesman and Zalapa2014). The 22 markers used here had the highest information index among those 43 markers. Overall, the 22 SSR markers detected 105 alleles in the SBRM-resistant germplasm and its parents (Table 3). The number of alleles per marker ranged from 2 to 8 and averaged 4.77 alleles per locus, while the number of effective alleles ranged from 1.18 to 5.61, with an average of 3.14 per locus. The information index of the markers ranged from 0.29 to 1.86 and averaged 1.21 per locus. Overall, the 22 markers detected 68 fewer alleles and 11 fewer effective alleles than in the previous study of eight divergent B. vulgaris genotypes. The low information indices and the low number of effective alleles for some of the loci in this study, most notably Unigene15403, Unigene16898, Unigene7492 and Unigene22354, compared to the previous study, documents the lower diversity and greater inbreeding that was expected among the genotypes of this study relative to the genotypes in the previous study. Except for PI 179180, all lines examined in this study were products of four USDA sugarbeet breeding programmes. Although the three samples from each line were unique genotypes, most lines were quite inbred and possessed few heterozygous loci (online Supplementary Table S1) resulting in reduced values for observed heterozygosity relative to expected heterozygosity for nearly every locus. Moreover, parental relationships between the SBRM-susceptible and SBRM-resistant germplasm and among all resistant germplasm, except for F1043, limit the genetic diversity among the genotypes in this study.
Genetic diversity statistics for 22 SSRs by locus based on 30 SBRM-resistant and -susceptible accessions
N, sample size; N a, number of alleles; N e, number of effective alleles; I, information index; H o, observed heterozygosity; H e, expected heterozygosity.
a Markers from Fugate et al. (Reference Fugate, Fajardo, Schlautman, Ferrareze, Bolton, Campbell, Wiesman and Zalapa2014).
The genetic diversity within SBRM-resistant and SBRM-susceptible accessions and within germplasm accessions was also determined (Table 4). Resistant germplasm included the four released populations (F1015, F1016, F1024 and F1043), the unreleased SBRM-resistant population, F1055 and the SBRM-resistant red beet, PI 179180. The number of alleles and the number of effective alleles for resistant germplasm were 3.86 and 2.64, respectively, while the information index for the resistant germplasm was 1.03. The SBRM-susceptible germplasm, comprised of F1010, FC-CLS, FC709 and C564, exhibited greater diversity than the resistant germplasm, with the number of alleles and the number of effective alleles at 4.23 and 2.92, respectively, and an information index of 1.14. Observed heterozygosity (H o) was lower than expected heterozygosity (H e) for both the resistant and susceptible germplasm, providing further evidence of the inbreeding of the germplasm in this study. The difference between H o and H e for the resistant germplasm, however, was greater than for the susceptible germplasm, suggesting that the resistant germplasm was more inbred than the susceptible germplasm. Genetic diversity also varied between individual accessions with the number of alleles between accessions ranging from 1.59 to 2.68, the number of effective alleles between accessions ranging from 1.40 to 2.32, and the information index between accessions ranging from 0.34 to 0.83. Among parental accessions, genetic diversity was greatest in F1010, the susceptible parent for F1015, F1016 and F1024, and was lowest in PI 179180, the resistant parent of F1043. Among SBRM-resistant releases, genetic diversity was greatest in F1015 and lowest in F1043.
Genetic diversity statistics based on 22 SSR loci by root maggot resistant/susceptible origin
N, sample size; N a, number of alleles; N e, number of effective alleles; I, information index; private, number of private alleles; H o, observed heterozygosity; H e, expected heterozygosity; RG, resistant germplasm; SP, susceptible parents.
Private alleles, that is, alleles present only within a defined group or accession, were found in resistant germplasm, susceptible germplasm, and seven of the 10 accessions (Table 4). Resistant and susceptible germplasm had 0.41 and 0.77 private alleles, respectively. Among accessions with private alleles, the number of private alleles ranged from 0.05 in PI 179180 to 0.32 in F1010.
Genotype clustering by PCoA indicated a tight grouping of three of the five SBRM-resistant germplasm lines, F1016, F1024 and F1055 (Fig. 1). Furthermore, an unweighted pair group method with arithmetic mean (UPGMA) dendrogram (Fig. 2) separated these same three SBRM-resistant germplasm lines from the susceptible parental lines, the other two resistant germplasm lines and the resistant accession with red globe-shaped roots (PI 179180). F1024 was selected from a population formed by crossing F1016 with a SBRM-susceptible line from USDA's Colorado breeding programme, FC-CLS. Both F1016 and F1024 were subjected to multiple cycles of mass selection followed by selection among half-sib families with a strong selection emphasis on SBRM resistance. Consequently, individual plants within each line are uniform, and root and canopy phenotypes of F1016 and F1024 are similar. F1055 was selected from a population formed by crossing F1024 with a SBRM-susceptible Rhizoctonia-resistant germplasm, FC709, developed by the USDA breeding programme in Ft. Collins, CO. F1016, F1024 and F1055 differ very little in their level of SBRM resistance (Table 2). Although there is no known close relationship between FC709 and FC-CLS, these two lines clustered together in both the PCoA and UPGMA dendrogram. However, both are SBRM-susceptible lines that were selected in a common environment and may have some common ancestry. Interestingly, FC709 and FC-CLS separated in PCoA and UPGMA from the F1016, F1024 and F1055 grouping. This suggests that the emphasis during selection on SBRM resistance eliminated much of the variation introduced to the original populations by the SBRM-susceptible parents and resulted in F1024 and F1055 being similar to F1016.
PCoA characterizing genetic relationships among five SBRM-resistant lines (RG; displayed in blue), four SBRM-susceptible accessions used as parental lines (SP; displayed in red) and one SBRM-resistant accession with red globe-shaped roots (RP; displayed in green), based upon SSR-markers.
UPGMA dendrogram summarizing genetic variation among five SBRM-resistant lines (displayed in blue), four SBRM-susceptible accessions used as parental lines (displayed in red) and one SBRM-resistant accession with red globe-shaped roots (displayed in green), based upon SSR-markers. Numbers represent branch confidence levels.
F1015 separated from the other SBRM-resistant germplasm in both the PCoA and the UPGMA dendrogram. F1015 was more closely associated with F1010, a heterogeneous broad-based SBRM-susceptible population and contributor to F1015, than to other SBRM-resistant germplasm that contained F1010 in their parentage (F1016, F1024 and F1055). F1015 retains some of the heterogeneity and vigour of F1010, and this heterogeneity and similarity to F1010 are confirmed by SSR marker analysis. However, SBRM damage ratings of F1015 are characteristically greater than those of the other SBRM-resistant lines, most likely due to genetic similarity with its susceptible parent (Table 2).
F1043 was identified as unique by UPGMA analysis when compared to the other four resistant germplasm and differed from other resistant germplasm by PCoA. F1043 is derived from a cross of the susceptible sugarbeet line, C564, and a red beet accession, PI 179180. The uniqueness of F1043 is perhaps expected since beet genotypes with red globe-shaped roots are not widely used in sugarbeet breeding programmes and all the genotypes in this study, except for PI 179180, could be characterized as cultivated sugarbeet germplasm. The uniqueness of F1043 holds the possibility that resistance for this genotype differs from that of the other four resistant germplasm. F1043, by UPGMA, was more closely related to its susceptible parent, C564, than were F1024 or F1055 with their susceptible parents. While SBRM resistance was a primary objective during the development of F1043, elimination of the red pigment and globe-shaped characteristics of the roots of the resistant parent, PI 179180, was also a factor. The dendrogram indicates that much of the variation of PI 179180 not related to SBRM resistance was not retained during the selection that resulted in F1043.
In both the PCoA and UPGMA dendrogram, the SBRM-susceptible parent F1010 demonstrated greater heterozygosity than other germplasm. F1010 was selected from a broad-based population formed by intermating biennial accessions from the USDA germplasm collection. These accessions represented diverse geographic areas and some non-sugarbeet types. This may explain the closer association of one of the F1010 (F1010-2) samples with the three PI 179180 samples than with other sugarbeet germplasm in the UPGMA dendrogram.
Conclusion
The objective of SBRM resistance breeding is to develop germplasm with sufficient resistance to be a useful component of an integrated SBRM management system. The SBRM-resistant germplasm lines generated to date provide sources for introducing SBRM resistance into elite breeding populations that also have the disease resistance and adaptation traits unique to specific production areas. Based upon their intertwined pedigrees, SBRM damage ratings (Table 2), and groupings by PCoA (Fig. 1) and UPGMA dendrogram (Fig. 2), F1016, F1024 and F1055 would likely be equally effective at introducing SBRM resistance into elite populations and would have similar impact, if any, on combining ability of the resulting elite parental lines. F1015 and F1043 are identified as alternatives to F1016, F1024 and F1055 as sources of SBRM resistance. Under severe SBRM infestations, the SBRM damage on F1015 is significantly less than that observed on susceptible germplasm or adapted hybrids, although the SBRM damage on F1015 is almost always greater than that observed on F1016, F1024 or F1055. The parental lines crossed to form the population from which F1043 was selected have no relationship with any of the SBRM resistant or susceptible lines contributing to other SBRM-resistant germplasm. F1043, therefore, has the potential to differ in the genes responsible for SBRM resistance. Unlike F1015, the low SBRM damage ratings recorded for F1043 are equal to those observed for F1016, F1024 and F1055. Overall, the results of this study, in conjunction with results from publications documenting the development of the germplasm lines (Campbell et al., Reference Campbell, Anderson and Dregseth2000, Reference Campbell, Panella and Smigocki2011; Campbell, Reference Campbell2017) and the performance of these lines when used as pollinators in testcross hybrids (Campbell et al., Reference Campbell, Miller, Rekoske and Smith2008; Campbell and Niehaus, Reference Campbell and Niehaus2008; Campbell et al., Reference Campbell, Panella and Smigocki2011; Campbell, Reference Campbell2018), provide information that will assist sugarbeet breeders in selecting SBRM-resistance sources to meet the unique objectives and constraints of individual sugarbeet breeding programmes.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262119000388.
Acknowledgements
The authors thank the Beet Sugar Development Foundation for partial financial support of this research. Mention of trade names or commercial products is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. USDA is an equal opportunity provider and employer.
